Selection, management, and outcome of vitamin K antagonist-treated patients with atrial fibrillation not switched to novel oral anticoagulants. Results from the Dresden NOAC registry.

Atrial fibrillation (AF) patients treated with well-controlled vitamin K antagonists (VKAs) may benefit less from non-vitamin K antagonist oral anticoagulants (NOACs) because they are supposed to be at low risk of thromboembolic and bleeding complications. However, little is known about the selection, management, and outcome of such "stable" VKA patients in current practice. We assessed characteristics, VKA persistence and 12 months' outcome of AF patients selected for VKA continuation. On March 1, 2013, the Dresden NOAC registry opened recruitment of patients continuing on VKA for sites that had been actively recruiting AF patients treated with NOACs in the prior 18 months. Patient characteristics were compared with those of NOAC patients from the same sites. Four hundred twenty-seven VKA patients had a significantly lower bleeding risk profile compared with 706 patients selected for NOAC treatment. For VKA, international normalised ratio time-in-therapeutic range before enrolment was 71 % and increased to 75 % during a mean follow-up of 15 months. Rates of stroke/transient ischaemic attack/systemic embolism were 1.3/100 patient-years (intention-to-treat) and 0.94/100 patient-years (as-treated). On-treatment rate of ISTH major bleeding was 4.15/100 patient-years (95 % CI 2.60-6.29) with a case-fatality rate of 16.3 % (all-cause mortality at day 90 after major bleeding). In conclusion, in daily care, AF patients selected for VKA therapy are healthier than those treated with NOAC, demonstrate a high quality of anticoagulant control and very low stroke rates. However, despite adequate patient selection and INR control, the risk of major VKA bleeding is unacceptably high and bleeding outcome is poor.